Ten prize-winning illusions

For some reason those of us here at WEIT—well, at least Matthew and I—are fascinated by optical illusions, crypsis, and other things that fool the eye. Well, we now have the Motherlode of Illusions: the ten 2016 finalists for Best Illusion of the Year Contest from the Neural Correlate Society.  You can see them all at the site, and should, but I’ll put my favorite three here along with the notes from the site:

Ambiguous Cylinder Illusion

– Kokichi Sugihara: “Ambiguous Cylinder Illusion”. Meiji University, Japan

The direct views of the objects and their mirror images generate quite different interpretations of the 3D shapes. They look like vertical cylinders, but their sections appear to be different; in one view they appear to be rectangles, while in the other view they appear to be circles. We cannot correct our interpretations although we logically know that they come from the same objects. Even if the object is rotated in front of a viewer, it is difficult to understand the true shape of the object, and thus the illusion does not disappear.

Didn’t help me much: I still don’t know what shape those damn things are!

A New Illusion at Your Elbow 

– Peter Brugger and Rebekka Meier:  “A New Illusion At Your Elbow”.  University Hospital Zurich, Switzerland

Move your finger slowly along a person’s forearm from the wrist towards the elbow crook – eyes closed, the person will anticipate touch in the elbow crook. This illusory anticipation may rest on our experience of tactile velocities that are usually much faster and make us believe we feel touch at a body location not yet reached. Neural characteristics of skin receptors specialized for slow motion may also contribute to the anticipation error. Like previously described illusions, the elbow crook illusion is larger on the non-dominant arm. Women showed a smaller illusion than men, confirming their reportedly superior cutaneous sensitivity.

You’ll surely want to try this yourself, but you can’t do it on yourself. Find a willing helper and report below. Malgorzata and I tried this on each other, and we both said “stop” 2-3 inches below the crook of the arm,

Remote controls:

– Arthur G. Shapiro : “Remote Controls ”. American University, USA

Two physically identical rectangular bars become light and dark at the same time, but in some conditions they look as if they wink in alternation.  The appearance of winking (alternating) or blinking (bars in sync) can be controlled by rectangles placed in the vicinity of the modulating bars: the bars blink when the rectangles are far away or adjacent to the bars but wink when there is a gap between the bars and the rectangles. The effect is remarkable because of the sudden change from wink to blink or vice versa, and because the change can occur across large distances.

h/t: David S.


  1. Alpha Neil
    Posted August 10, 2016 at 1:29 pm | Permalink

    The “remote control” illusion didn’t work as well for me. They looked out of phase all the time unless I intentionally put both bars into my peripheral vision. Maybe that’s the key to the illusion.

    The cylinders are very cool and (I think) a good metaphor for ideology – perspective can make things look like something they are not.

    • Heather Hastie
      Posted August 10, 2016 at 4:20 pm | Permalink

      I love optical illusions like these. However, the remote control one didn’t work for me either. They were closer to being in phase, but not completely.

      • HaggisForBrains
        Posted August 11, 2016 at 11:21 am | Permalink

        Me too.

    • sshort
      Posted August 10, 2016 at 5:03 pm | Permalink

      Having seen similar illusions, and since i was viewing on an iPad, i knew to toggle the video to full screen, pull the iPad a bit closer to my face (so that it captured most of my field of vision) and kept focused on the centerline. Then it works quite well.

      Not completely necessary for the effect, but it helps.

    • sshort
      Posted August 10, 2016 at 5:08 pm | Permalink

      Having seen illusions like remote contol before, i knew to toggle the vid to full screen, pull my iPad a bit closer to my face so it occupied most of my field of vision, and focused on the centerline. Then it works quite well.

      Not completely neccessary for the effect, but it helps.

      • infiniteimprobabilit
        Posted August 10, 2016 at 6:34 pm | Permalink

        I found that worked best for me too.

        However, downloading the video and running it 1/4 speed in VLC, I can confirm that the bars are blinking ‘in phase’ the whole time. That is, they go bright, then dark, simultaneously.

        However, the fact one is on dark background and one is on bright results in opposed conflicting visual cues, as one merges into the background exactly when the other stands out most – so they appear out of phase. The fringing bars (which are ‘opposite’ to the background) presumably modulate this effect.


    • Posted August 11, 2016 at 1:30 am | Permalink

      At first I thought perhaps the remote control one didn’t work well for me because I have monocular vision. But I’m guessing that with this many people saying it didn’t work well, that’s probably not (much of) a factor.

  2. Sastra
    Posted August 10, 2016 at 1:35 pm | Permalink

    “I’m sorry, I know what I experienced.”

    Heh. I love these things.

  3. Posted August 10, 2016 at 1:42 pm | Permalink

    Here is an explanation of how the ambiguous cylinders were done, from Captain Disillusion:

    • WT
      Posted August 10, 2016 at 2:57 pm | Permalink

      Beat me to it. He’s got some great videos.

    • Mark Sturtevant
      Posted August 10, 2016 at 5:17 pm | Permalink

      That was a very well done explanation. Thanks for that!

  4. Posted August 10, 2016 at 1:43 pm | Permalink

    I like the Gabors the best.


  5. Stephen Barnard
    Posted August 10, 2016 at 1:51 pm | Permalink

    I love the first one.

    I worked in computer vision (a branch of artificial intelligence) for many years and made some modest contributions to the field. I was interested in illusions for their potential to reveal the workings of the human visual system (HVS).

    The HVS constructs an interpretation of reality from insufficient evidence. Three dimensional vision is what is called in mathematics an ill-posed problem: a problem with more than one solution, that depends discontinuously on(insufficient) data. That is the source of many interesting illusions.

    That the HVS can construct accurate interpretations in nearly all real-world situations is remarkable, and a testament to the power of evolution.

    I left the field because it was too hard. No one has the faintest idea how the HVS works, except for very low-level processes.

  6. Mark Sturtevant
    Posted August 10, 2016 at 2:41 pm | Permalink

    Cool! The Millusion illusion at the web site was also noteworthy. I found that I could make the windmills rotate one way, and then reverse their direction. This was done by focusing on the blades as they come toward me, or focusing on the blades as they swing away from me.

    • infiniteimprobabilit
      Posted August 11, 2016 at 3:31 am | Permalink

      I’ve experienced that reversal of rotation with real wind turbines. Sometimes in poor light it’s impossible to tell which way they’re turning and my mind flip-flops from one direction of rotation to the other as if attempting to guess.


      • infiniteimprobabilit
        Posted August 11, 2016 at 3:39 am | Permalink

        P.S. There’s another illusion seen with the (much rarer) large two-blade turbines – when seen from nearly side-on, as the blade comes to line up with your direction of view it appears to rocket past you, then it slows down for most of its half-turn, then rockets past you again.


  7. Torbjörn Larsson
    Posted August 10, 2016 at 4:11 pm | Permalink

    Like Alpha Neil, the “Remote Controls” illusion doesn’t work for me.

    And the “Elbow illusion” works so-so at 0-2 cm, maybe because I made a series of quick test on myself. (It is good to “know” your actual arm and under arm length…)

    The “Ambiguous cylinder” illusion works well though. And I grokked the actual cross section from the rotations shown in the original video. But I thought the trick was merely that the upper surface was at an angle, not the curvature shown in the thread video. Physics fail.

    • Torbjörn Larsson
      Posted August 10, 2016 at 4:12 pm | Permalink

      Oy. Underarm; it is good to know biology terms too…

    • infiniteimprobabilit
      Posted August 10, 2016 at 6:40 pm | Permalink

      I think the cross-sectional curvature (a compromise between a square and a circle) is essential, so that the top edge can be cut to show both a square and a circle. I think, if the cylinder was circular, the top edge could still be cut to show a square, but the shape of the top edge would then be too extreme to ever appear as a circle from the opposite direction. Similarly if the cylinder was square, with the top edge cut to show a circle, it couldn’t then ever appear square.


  8. Posted August 10, 2016 at 4:41 pm | Permalink

    Reblogged this on The Logical Place.

  9. Posted August 10, 2016 at 4:51 pm | Permalink

    I’m surprised to see the Elbow illusion winning the 2016 award. I remember something similar at school back in the 1980’s (or earlier), although we ‘ran’ up the elbow with two fingers, rather than just sliding one.

  10. infiniteimprobabilit
    Posted August 10, 2016 at 6:12 pm | Permalink

    In the top illusion, they must obviously be cylinders of some sort, with the ends (top and bottom) cut in a cunning curve.

    Further googling shows they’re actually squarish cylinders with rounded corners.

    But very difficult to visualise even when you ‘know’ the answer.


  11. Ken Pidcock
    Posted August 10, 2016 at 7:43 pm | Permalink

    I’ve long been particularly fond of these. More detail can be found in this article from the May-June 2002 issue of American Scientist. The cover of that issue, which references the article, is just breathtaking, but I can’t find a good image of it.

  12. Posted August 10, 2016 at 8:24 pm | Permalink

    I am Arthur Shapiro creator of Remote Controls and fan of WEIT. Thanks for the useful feedback about the video. The demo works beautifully in the lab and for my colleagues/students in my hallway, but clearly I should have done some A/B testing prior to the contest. Alas. Glad you like the other illusions, though! It is a fun contest.

    • infiniteimprobabilit
      Posted August 11, 2016 at 3:09 am | Permalink

      Hey Arthur, it works pretty well on the whole. Do not be discouraged by this bunch of hyper-critical commenters (and I include myself in that). If you gave us a Faberge egg we’d start picking it apart to see how it’s made.


One Trackback/Pingback

  1. […] via Ten prize-winning illusions — Why Evolution Is True […]

%d bloggers like this: